The Saturday freeze call that goes to voicemail is gone by Monday
Winter demand arrives in bursts, at the wrong hours, and it does not wait for your office to open.
A sustained cold snap produces days of calls compressed into hours, often in the evening and over a weekend. Whoever picks up first tends to get the measure, because the caller has three other numbers on the screen.
Plan coverage for the freeze weeks specifically rather than for an average week. An overflow answering service that takes a real intake, and an automatic text back within minutes when a call is missed, cover most of the gap.
A returned call the next business day is not the same as an answered call. Track answer rate and missed calls by hour so you can see exactly which hours you are giving away.
Give the after-hours script the same qualification questions as the daytime one. An unqualified message is only slightly better than a missed call.
Three questions on the first call separate a two-flat owner from a single stuck sash
Qualification on the phone is cheaper than qualification on somebody's porch.
Ask the building type, the number of openings and who owns the building. A greystone owner replacing fourteen openings and a tenant with a stuck sash sound similar for the first thirty seconds and are not the same business at all.
Ask about access next. Rear alley, gangway, street parking, occupied units, and what floor. Access determines crew size and the length of the day, so it belongs in the conversation before a date is offered.
Give small jobs a cheaper path rather than refusing them. A two-opening inquiry that gets a phone quote and a photo review keeps the estimator's day free for whole-building work.
Write the script down and hold to it. The value comes from asking the same questions every time, so the answers become data instead of impressions.
Winter measures cancel, so confirm twice and keep the slot warm
The appointment you booked in a freeze week is the one most likely to evaporate.
Snow, a work schedule change or a homeowner simply cooling off between the call and the visit all take measures off the calendar. Cancellations are ordinary here rather than a sign of a bad lead.
Confirm at booking and again the morning of, by text as well as phone. Offer to rebook in the same message rather than waiting for the homeowner to call back.
Track held measures separately from booked measures. Two shops with identical booking numbers can have very different revenue, and the gap between the two is where the money sits.
Keep a short standby list. When a Berwyn appointment drops at nine in the morning, a nearby inquiry that has been waiting three days can often fill the slot the same day.
Follow-up should be built around the install calendar, not a seven day cadence
A quote given in February gets decided when the ground thaws, and a generic drip sequence will have given up by then.
Whole-house replacement is a considered purchase with two or three competing bids and, here, a seasonal decision point. A homeowner who priced work during a cold snap often signs when spring scheduling opens.
Build the sequence to outlast that gap. Fewer, more useful touches over months beat five emails in two weeks. A note when spring slots open is more welcome than a fourth reminder about the same quote.
Segment by what they were quoting. A landlord phasing a three-flat across two seasons needs a different rhythm from a homeowner replacing six windows at once.
Make sure somebody owns the list. Automation sends the message, but the call that gets a signature is usually placed by a person looking at a name they recognize.
One close rate averages the emergency call and the planned project into nonsense
Your close rate is at least three different numbers wearing one label.
The winter emergency inquiry, the planned whole-house project and the multi-unit building owner close at different rates on different timelines. Averaged together, they describe a business that does not exist.
Report by inquiry type, by county and by season. A dashboard that stops at leads and cost per lead hides the two places most dealers lose money, which are unanswered calls and canceled measures.
Count signed openings alongside signed jobs. A fourteen opening two-flat and a single patio door look identical in a lead count and nothing alike in revenue.
Instrument the whole path: inquiry, answered, qualified, booked, held, quoted, signed. Missing any one stage means you are guessing about which stage is broken.
Run one change per season and let the first freeze week grade it
One metro does not generate enough traffic for small split tests, so stop trying to run them.
Button colors and headline variants need volume you do not have. Attempting them produces results that look real and are noise, and acting on noise is worse than doing nothing.
Change big things instead, one at a time, and read them in booked and held measures rather than in clicks. After-hours coverage, the qualification script, the follow-up cadence, the quote document itself.
Use the seasons as your read. A change made in October is graded by the freeze weeks. A change made in March is graded by spring bookings. Keep a dated log so nobody argues about what was running when.
Call recordings are your best qualitative evidence. Listening to twenty calls from a busy week usually surfaces the leak faster than any report, and it costs an afternoon.
Questions we actually get
- What should we fix first?
- Almost always phone coverage during cold weather, because it is the largest leak and the cheapest to close. Look at answer rate by hour and by day across your last freeze week. If calls are being missed in the evening or on weekends, no amount of website work will matter until that is handled.
- How do we measure this without an expensive setup?
- Call tracking on the phone number, a lead identifier passed from the form into your CRM, and stage fields for booked, held, quoted and signed. Most of the value comes from counting held measures and signed openings, which almost nobody does and which usually reveals the problem within a month.
- Can you A/B test our site?
- Not usefully at one metro's volume, and we will not pretend otherwise. Small effect tests need traffic you do not have, and a result that looks significant on this sample generally is not. We change larger things sequentially and read them in booked and held measures instead.
- Is a text-back on missed calls worth setting up?
- Generally yes for this trade, because winter callers are contacting several companies in a short window. An immediate text that acknowledges the call and offers a booking option keeps you in the conversation while a voicemail you return tomorrow often does not.
- How long should we keep following up on an unsigned quote?
- Longer than most shops do, and around the install calendar rather than on a weekly timer. A quote from a January cold snap can turn into spring work. Fewer, better timed touches over several months tend to recover more than an aggressive two week sequence.
